Google Play Review & Competitor Intelligence

Turn public Google Play reviews into structured product intelligence: sentiment, complaint categories, feature requests, questions, purchase intent, competitor mentions, switching signals, app-version trends, app metadata, and change monitoring.

Why this Actor

Raw Google Play review scrapers are common. This Actor is designed for product, growth, support, and competitive-intelligence workflows that need normalized signals rather than only review text.

Source and stability

The default source is the public Google Play web review corpus through the pinned app-reviews==0.6.0 package. No Google account, customer API key, login, or browser is required. Google Play's public web endpoint is undocumented and rate-limited, so it can change without notice. The Actor therefore uses bounded request budgets, retries/backoff, direct-first networking, optional proxy/session rotation, cycle-safe pagination, diagnostics, and a circuit breaker. It never attempts to solve CAPTCHA or bypass authentication.

Google Play reviews are a global corpus in this source; a country selector would imply precision the source does not provide, so the current version intentionally does not expose one. It also uses newest-first collection only to keep monitoring semantics stable.

Input

• apps: package names or Play Store URLs.
• maxReviewsPerApp: hard per-app result cap.
• ratings: optional 1-5 star filter.
• competitorNames: names used for mention/switching analysis.
• includeReviewerName: off by default to minimize unnecessary personal data.
• includeAppMetadata: fetch app/developer/category/version/rating/price metadata.
• monitorChanges: actor-scoped baseline/change tracking.
• onlyNewReviews: after a baseline exists, emit only new/changed reviews.
• cacheTtlMinutes, forceRefresh: non-monitoring cache controls.
• bounded concurrency/request/retry/timeout/proxy controls.

Output

Each review row includes the normalized review plus:

• sentiment
• complaintCategories and complaintScore
• featureRequest
• questionSignal
• purchaseIntent
• competitorsMentioned
• switchingSignal, switchingScore, switchingPhrases
• audienceIntent
• baseline-safe monitoring flags

Each app also gets an app_summary with rating distribution, average rating in the run, sentiment counts, top complaint categories, competitor mentions, app-version counts, and monitoring counts.

The built-in semantic heuristics are strongest for English-language review text. Raw review/rating extraction is not language-restricted.

Monitoring behavior

The first monitoring run establishes a baseline; reviews are not falsely labeled new. Later runs compare the same scoped configuration. Cache/state keys include inputs that materially affect results so changing limits, ratings, metadata inclusion, or competitor names does not silently reuse an incompatible cache/state scope.

Pricing

This Actor uses Pay Per Event pricing with the primary event review-returned at $0.0005 per useful chargeable review ($0.50 per 1,000 reviews).

Summaries, diagnostics, retries, proxy rotations, cached rows, duplicates, and unchanged monitoring rows are not charged as review events.

Limits

• Public Google Play web scraping is unofficial and may change or rate-limit.
• Reviewer country is not inferred.
• Reviewer display names are optional and disabled by default.
• Developer replies / thumbs-up are not currently exposed because the normalized dependency API does not guarantee those fields.
• CAPTCHA is never solved.